Skip to contentSkip to conversation

Enable vdpa net migration with features depending on CVQ

RFC PATCH In review Eugenio Pérez 3 participants 16 messages Jul 6, 2023 – Jul 20, 2023 Networking

lore
Share & tools
lore
#0 lore raw

cover Enable vdpa net migration with features depending on CVQ

At this moment the migration of net features that depends on CVQ is not possible, as there is no reliable way to restore the device state like mac address, number of enabled queues, etc to the destination. This is mainly caused because the device must only read CVQ, and process all the commands before resuming the dataplane.

This RFC lift that requirement, sending the VHOST_VDPA_SET_VRING_ENABLE ioctl for dataplane vqs only after the device has processed all commands. If this method is valid or not, or if it must be signalled by the parent driver somehow, is still under discussion. In case it is valid, this code allows testing the vDPA device for it.

Eugenio Pérez (6):
vdpa: export vhost_vdpa_set_vring_ready
vdpa: add should_enable op
vdpa: use virtio_ops->should_enable at vhost_vdpa_set_vrings_ready
vdpa: add stub vhost_vdpa_should_enable
vdpa: delay enable of data vqs
vdpa: remove net cvq migration blocker

 include/hw/virtio/vhost-vdpa.h |  9 +++++++
 hw/virtio/vhost-vdpa.c         | 33 +++++++++++++++++++------
 net/vhost-vdpa.c               | 45 +++++++++++++++++++++++++---------
 hw/virtio/trace-events         |  2 +-
 4 files changed, 68 insertions(+), 21 deletions(-)
Signature · 1 line
2.39.3
#1 lore raw

RFC PATCH 1/6 vdpa: export vhost_vdpa_set_vring_ready

The vhost-vdpa net backend needs to enable vrings in a different order than default, so export it.

No functional change intended except for tracing, that now includes the (virtio) index being enabled and the return value of the ioctl.

Still ignoring return value of this function if called from vhost_vdpa_dev_start, as reorganize calling code around it is out of the scope of this series.

---
 include/hw/virtio/vhost-vdpa.h |  1 +
 hw/virtio/vhost-vdpa.c         | 28 ++++++++++++++++++++--------
 hw/virtio/trace-events         |  2 +-
 3 files changed, 22 insertions(+), 9 deletions(-)
include/hw/virtio/vhost-vdpa.h +1 -0
index e64bfc7f98..5407d54fd7 100644
@@ -57,6 +57,7 @@ typedef struct vhost_vdpa {
 } VhostVDPA;
 
 int vhost_vdpa_get_iova_range(int fd, struct vhost_vdpa_iova_range *iova_range);
+int vhost_vdpa_set_vring_ready(struct vhost_vdpa *v, unsigned idx);
 
 int vhost_vdpa_dma_map(struct vhost_vdpa *v, uint32_t asid, hwaddr iova,
                        hwaddr size, void *vaddr, bool readonly);
hw/virtio/vhost-vdpa.c +20 -8
index 3c575a9a6e..5978d970ee 100644
@@ -528,6 +528,19 @@ int vhost_vdpa_get_iova_range(int fd, struct vhost_vdpa_iova_range *iova_range)
     return ret < 0 ? -errno : 0;
 }
 
+int vhost_vdpa_set_vring_ready(struct vhost_vdpa *v, unsigned idx)
+{
+    struct vhost_dev *dev = v->dev;
+    struct vhost_vring_state state = {
+        .index = idx,
+        .num = 1,
+    };
+    int r = vhost_vdpa_call(dev, VHOST_VDPA_SET_VRING_ENABLE, &state);
+
+    trace_vhost_vdpa_set_vring_ready(dev, idx, r);
+    return r;
+}
+
 /*
  * The use of this function is for requests that only need to be
  * applied once. Typically such request occurs at the beginning
@@ -872,16 +885,15 @@ static int vhost_vdpa_get_vq_index(struct vhost_dev *dev, int idx)
     return idx;
 }
 
-static int vhost_vdpa_set_vring_ready(struct vhost_dev *dev)
+static int vhost_vdpa_set_vrings_ready(struct vhost_dev *dev)
 {
+    struct vhost_vdpa *v = dev->opaque;
     int i;
-    trace_vhost_vdpa_set_vring_ready(dev);
+
+    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_VDPA);
+
     for (i = 0; i < dev->nvqs; ++i) {
-        struct vhost_vring_state state = {
-            .index = dev->vq_index + i,
-            .num = 1,
-        };
-        vhost_vdpa_call(dev, VHOST_VDPA_SET_VRING_ENABLE, &state);
+        vhost_vdpa_set_vring_ready(v, dev->vq_index + i);
     }
     return 0;
 }
@@ -1294,7 +1306,7 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
         if (unlikely(!ok)) {
             return -1;
         }
-        vhost_vdpa_set_vring_ready(dev);
+        vhost_vdpa_set_vrings_ready(dev);
     } else {
         vhost_vdpa_suspend(dev);
         vhost_vdpa_svqs_stop(dev);
hw/virtio/trace-events +1 -1
index 8f8d05cf9b..4f6a6ba428 100644
@@ -46,7 +46,7 @@ vhost_vdpa_set_features(void *dev, uint64_t features) "dev: %p features: 0x%"PRI
 vhost_vdpa_get_device_id(void *dev, uint32_t device_id) "dev: %p device_id %"PRIu32
 vhost_vdpa_reset_device(void *dev, uint8_t status) "dev: %p status: 0x%"PRIx8
 vhost_vdpa_get_vq_index(void *dev, int idx, int vq_idx) "dev: %p idx: %d vq idx: %d"
-vhost_vdpa_set_vring_ready(void *dev) "dev: %p"
+vhost_vdpa_set_vring_ready(void *dev, unsigned i, int r) "dev: %p, idx: %u, r: %d"
 vhost_vdpa_dump_config(void *dev, const char *line) "dev: %p %s"
 vhost_vdpa_set_config(void *dev, uint32_t offset, uint32_t size, uint32_t flags) "dev: %p offset: %"PRIu32" size: %"PRIu32" flags: 0x%"PRIx32
 vhost_vdpa_get_config(void *dev, void *config, uint32_t config_len) "dev: %p config: %p config_len: %"PRIu32
Signature · 1 line
2.39.3
#2 lore raw

Re: [RFC PATCH 1/6] vdpa: export vhost_vdpa_set_vring_ready

Quoted 56 lines · Eugenio Pérez

The vhost-vdpa net backend needs to enable vrings in a different order than default, so export it.

No functional change intended except for tracing, that now includes the (virtio) index being enabled and the return value of the ioctl.

Still ignoring return value of this function if called from vhost_vdpa_dev_start, as reorganize calling code around it is out of the scope of this series.

---
 include/hw/virtio/vhost-vdpa.h |  1 +
 hw/virtio/vhost-vdpa.c         | 28 ++++++++++++++++++++--------
 hw/virtio/trace-events         |  2 +-
 3 files changed, 22 insertions(+), 9 deletions(-)
include/hw/virtio/vhost-vdpa.h +1 -0
index e64bfc7f98..5407d54fd7 100644
@@ -57,6 +57,7 @@ typedef struct vhost_vdpa {
 } VhostVDPA;
 
 int vhost_vdpa_get_iova_range(int fd, struct vhost_vdpa_iova_range *iova_range);
+int vhost_vdpa_set_vring_ready(struct vhost_vdpa *v, unsigned idx);
 
 int vhost_vdpa_dma_map(struct vhost_vdpa *v, uint32_t asid, hwaddr iova,
                        hwaddr size, void *vaddr, bool readonly);
hw/virtio/vhost-vdpa.c +13 -0
index 3c575a9a6e..5978d970ee 100644
@@ -528,6 +528,19 @@ int vhost_vdpa_get_iova_range(int fd, struct vhost_vdpa_iova_range *iova_range)
     return ret < 0 ? -errno : 0;
 }
 
+int vhost_vdpa_set_vring_ready(struct vhost_vdpa *v, unsigned idx)
+{
+    struct vhost_dev *dev = v->dev;
+    struct vhost_vring_state state = {
+        .index = idx,
+        .num = 1,
+    };
+    int r = vhost_vdpa_call(dev, VHOST_VDPA_SET_VRING_ENABLE, &state);
+
+    trace_vhost_vdpa_set_vring_ready(dev, idx, r);
+    return r;
+}
+
 /*
  * The use of this function is for requests that only need to be
  * applied once. Typically such request occurs at the beginning
@@ -872,16 +885,15 @@ static int vhost_vdpa_get_vq_index(struct vhost_dev *dev, int idx)
     return idx;

hw/virtio/vhost-vdpa.c static int vhost_vdpa_get_vq_index(struct vhost_dev *dev, int idx) · line 893

Quoted 10 lines
hw/virtio/vhost-vdpa.c +4 -2
 }
 
-static int vhost_vdpa_set_vring_ready(struct vhost_dev *dev)
+static int vhost_vdpa_set_vrings_ready(struct vhost_dev *dev)
 {
+    struct vhost_vdpa *v = dev->opaque;
     int i;
-    trace_vhost_vdpa_set_vring_ready(dev);
+
+    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_VDPA);

Nit: any reason we need to add this assert in this patch?

Thanks

Quoted 36 lines
hw/virtio/vhost-vdpa.c +3 -6
+
     for (i = 0; i < dev->nvqs; ++i) {
-        struct vhost_vring_state state = {
-            .index = dev->vq_index + i,
-            .num = 1,
-        };
-        vhost_vdpa_call(dev, VHOST_VDPA_SET_VRING_ENABLE, &state);
+        vhost_vdpa_set_vring_ready(v, dev->vq_index + i);
     }
     return 0;
 }
@@ -1294,7 +1306,7 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
         if (unlikely(!ok)) {
             return -1;
         }
-        vhost_vdpa_set_vring_ready(dev);
+        vhost_vdpa_set_vrings_ready(dev);
     } else {
         vhost_vdpa_suspend(dev);
         vhost_vdpa_svqs_stop(dev);
hw/virtio/trace-events +1 -2
index 8f8d05cf9b..4f6a6ba428 100644
@@ -46,7 +46,7 @@ vhost_vdpa_set_features(void *dev, uint64_t features) "dev: %p features: 0x%"PRI
 vhost_vdpa_get_device_id(void *dev, uint32_t device_id) "dev: %p device_id %"PRIu32
 vhost_vdpa_reset_device(void *dev, uint8_t status) "dev: %p status: 0x%"PRIx8
 vhost_vdpa_get_vq_index(void *dev, int idx, int vq_idx) "dev: %p idx: %d vq idx: %d"
-vhost_vdpa_set_vring_ready(void *dev) "dev: %p"
+vhost_vdpa_set_vring_ready(void *dev, unsigned i, int r) "dev: %p, idx: %u, r: %d"
 vhost_vdpa_dump_config(void *dev, const char *line) "dev: %p %s"
 vhost_vdpa_set_config(void *dev, uint32_t offset, uint32_t size, uint32_t flags) "dev: %p offset: %"PRIu32" size: %"PRIu32" flags: 0x%"PRIx32
 vhost_vdpa_get_config(void *dev, void *config, uint32_t config_len) "dev: %p config: %p config_len: %"PRIu32
--

2.39.3

#3 lore raw

Re: [RFC PATCH 1/6] vdpa: export vhost_vdpa_set_vring_ready

Quoted 60 lines · Jason Wang
Quoted 58 lines · Eugenio Pérez

The vhost-vdpa net backend needs to enable vrings in a different order than default, so export it.

No functional change intended except for tracing, that now includes the (virtio) index being enabled and the return value of the ioctl.

Still ignoring return value of this function if called from vhost_vdpa_dev_start, as reorganize calling code around it is out of the scope of this series.

---
 include/hw/virtio/vhost-vdpa.h |  1 +
 hw/virtio/vhost-vdpa.c         | 28 ++++++++++++++++++++--------
 hw/virtio/trace-events         |  2 +-
 3 files changed, 22 insertions(+), 9 deletions(-)
include/hw/virtio/vhost-vdpa.h +1 -0
index e64bfc7f98..5407d54fd7 100644
@@ -57,6 +57,7 @@ typedef struct vhost_vdpa {
 } VhostVDPA;
 
 int vhost_vdpa_get_iova_range(int fd, struct vhost_vdpa_iova_range *iova_range);
+int vhost_vdpa_set_vring_ready(struct vhost_vdpa *v, unsigned idx);
 
 int vhost_vdpa_dma_map(struct vhost_vdpa *v, uint32_t asid, hwaddr iova,
                        hwaddr size, void *vaddr, bool readonly);
hw/virtio/vhost-vdpa.c +13 -0
index 3c575a9a6e..5978d970ee 100644
@@ -528,6 +528,19 @@ int vhost_vdpa_get_iova_range(int fd, struct vhost_vdpa_iova_range *iova_range)
     return ret < 0 ? -errno : 0;
 }
 
+int vhost_vdpa_set_vring_ready(struct vhost_vdpa *v, unsigned idx)
+{
+    struct vhost_dev *dev = v->dev;
+    struct vhost_vring_state state = {
+        .index = idx,
+        .num = 1,
+    };
+    int r = vhost_vdpa_call(dev, VHOST_VDPA_SET_VRING_ENABLE, &state);
+
+    trace_vhost_vdpa_set_vring_ready(dev, idx, r);
+    return r;
+}
+
 /*
  * The use of this function is for requests that only need to be
  * applied once. Typically such request occurs at the beginning
@@ -872,16 +885,15 @@ static int vhost_vdpa_get_vq_index(struct vhost_dev *dev, int idx)
     return idx;
 }
 
Quoted 11 lines

hw/virtio/vhost-vdpa.c static int vhost_vdpa_get_vq_index(struct vhost_dev *dev, int idx) · line 893

Quoted 8 lines
hw/virtio/vhost-vdpa.c +4 -2
-static int vhost_vdpa_set_vring_ready(struct vhost_dev *dev)
+static int vhost_vdpa_set_vrings_ready(struct vhost_dev *dev)
 {
+    struct vhost_vdpa *v = dev->opaque;
     int i;
-    trace_vhost_vdpa_set_vring_ready(dev);
+
+    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_VDPA);

Nit: any reason we need to add this assert in this patch?

Other vhost_ops asserts, but sure we can leave it out.

Thanks!

Quoted 39 lines

Thanks

Quoted 36 lines
hw/virtio/vhost-vdpa.c +3 -6
+
     for (i = 0; i < dev->nvqs; ++i) {
-        struct vhost_vring_state state = {
-            .index = dev->vq_index + i,
-            .num = 1,
-        };
-        vhost_vdpa_call(dev, VHOST_VDPA_SET_VRING_ENABLE, &state);
+        vhost_vdpa_set_vring_ready(v, dev->vq_index + i);
     }
     return 0;
 }
@@ -1294,7 +1306,7 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
         if (unlikely(!ok)) {
             return -1;
         }
-        vhost_vdpa_set_vring_ready(dev);
+        vhost_vdpa_set_vrings_ready(dev);
     } else {
         vhost_vdpa_suspend(dev);
         vhost_vdpa_svqs_stop(dev);
hw/virtio/trace-events +1 -2
index 8f8d05cf9b..4f6a6ba428 100644
@@ -46,7 +46,7 @@ vhost_vdpa_set_features(void *dev, uint64_t features) "dev: %p features: 0x%"PRI
 vhost_vdpa_get_device_id(void *dev, uint32_t device_id) "dev: %p device_id %"PRIu32
 vhost_vdpa_reset_device(void *dev, uint8_t status) "dev: %p status: 0x%"PRIx8
 vhost_vdpa_get_vq_index(void *dev, int idx, int vq_idx) "dev: %p idx: %d vq idx: %d"
-vhost_vdpa_set_vring_ready(void *dev) "dev: %p"
+vhost_vdpa_set_vring_ready(void *dev, unsigned i, int r) "dev: %p, idx: %u, r: %d"
 vhost_vdpa_dump_config(void *dev, const char *line) "dev: %p %s"
 vhost_vdpa_set_config(void *dev, uint32_t offset, uint32_t size, uint32_t flags) "dev: %p offset: %"PRIu32" size: %"PRIu32" flags: 0x%"PRIx32
 vhost_vdpa_get_config(void *dev, void *config, uint32_t config_len) "dev: %p config: %p config_len: %"PRIu32
--

2.39.3

#4 lore raw

RFC PATCH 2/6 vdpa: add should_enable op

In reply to Eugenio Pérez

To restore the device at the destination of a live migration we send the commands through control virtqueue. For a device to read CVQ it must have received the DRIVER_OK status bit.

However this opens a window where the device could start receiving packets in rx queue 0 before it receives the RSS configuration. To avoid that, we will not send vring_enable until all configuration is used by the device.

As a first step, enable a new vitio ops per vhost_vdpa device to know if we should enable a virtqueue or not. This srtuct can be reused in the future to add more actions to vhost_vdpa that depend on the virtIO kind of device.

---
 include/hw/virtio/vhost-vdpa.h | 8 ++++++++
 1 file changed, 8 insertions(+)
include/hw/virtio/vhost-vdpa.h +8 -0
index 5407d54fd7..3d330d439a 100644
@@ -30,6 +30,13 @@ typedef struct VhostVDPAHostNotifier {
     void *addr;
 } VhostVDPAHostNotifier;
 
+struct vhost_vdpa;
+typedef bool (*vhost_vdpa_virtio_should_enable_op)(const struct vhost_vdpa *v);
+
+typedef struct VhostVDPAVirtIOOps {
+    vhost_vdpa_virtio_should_enable_op should_enable;
+} VhostVDPAVirtIOOps;
+
 typedef struct vhost_vdpa {
     int device_fd;
     int index;
@@ -48,6 +55,7 @@ typedef struct vhost_vdpa {
     VhostIOVATree *iova_tree;
     GPtrArray *shadow_vqs;
     const VhostShadowVirtqueueOps *shadow_vq_ops;
+    const VhostVDPAVirtIOOps *virtio_ops;
     void *shadow_vq_ops_opaque;
     struct vhost_dev *dev;
     Error *migration_blocker;
Signature · 1 line
2.39.3
#5 lore raw

Re: [RFC PATCH 2/6] vdpa: add should_enable op

Quoted 7 lines · Eugenio Pérez

To restore the device at the destination of a live migration we send the commands through control virtqueue. For a device to read CVQ it must have received the DRIVER_OK status bit.

However this opens a window where the device could start receiving packets in rx queue 0 before it receives the RSS configuration.

I think this can also happen in the case of automatic steering.

Quoted 6 lines

To
avoid that, we will not send vring_enable until all configuration is used by the device.

As a first step, enable a new vitio ops per vhost_vdpa device to know if we should enable a virtqueue or not. This srtuct can be reused in the

typo.

Thanks

Quoted 37 lines

future to add more actions to vhost_vdpa that depend on the virtIO kind of device.

---
 include/hw/virtio/vhost-vdpa.h | 8 ++++++++
 1 file changed, 8 insertions(+)
include/hw/virtio/vhost-vdpa.h +8 -1
index 5407d54fd7..3d330d439a 100644
@@ -30,6 +30,13 @@ typedef struct VhostVDPAHostNotifier {
     void *addr;
 } VhostVDPAHostNotifier;
 
+struct vhost_vdpa;
+typedef bool (*vhost_vdpa_virtio_should_enable_op)(const struct vhost_vdpa *v);
+
+typedef struct VhostVDPAVirtIOOps {
+    vhost_vdpa_virtio_should_enable_op should_enable;
+} VhostVDPAVirtIOOps;
+
 typedef struct vhost_vdpa {
     int device_fd;
     int index;
@@ -48,6 +55,7 @@ typedef struct vhost_vdpa {
     VhostIOVATree *iova_tree;
     GPtrArray *shadow_vqs;
     const VhostShadowVirtqueueOps *shadow_vq_ops;
+    const VhostVDPAVirtIOOps *virtio_ops;
     void *shadow_vq_ops_opaque;
     struct vhost_dev *dev;
     Error *migration_blocker;
--

2.39.3

#6 lore raw

RFC PATCH 3/6 vdpa: use virtio_ops->should_enable at vhost_vdpa_set_vrings_ready

In reply to Eugenio Pérez

This allow to skip some rings that qemu does not want to enable.

---
 hw/virtio/vhost-vdpa.c | 5 +++++
 1 file changed, 5 insertions(+)
hw/virtio/vhost-vdpa.c +5 -0
index 5978d970ee..f51f5d9969 100644
@@ -892,6 +892,11 @@ static int vhost_vdpa_set_vrings_ready(struct vhost_dev *dev)
 
     assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_VDPA);
 
+    if (v->virtio_ops && v->virtio_ops->should_enable
+        && !(v->virtio_ops->should_enable(v))) {
+        return 0;
+    }
+
     for (i = 0; i < dev->nvqs; ++i) {
         vhost_vdpa_set_vring_ready(v, dev->vq_index + i);
     }
Signature · 1 line
2.39.3
#7 lore raw

RFC PATCH 4/6 vdpa: add stub vhost_vdpa_should_enable

In reply to Eugenio Pérez

To restore the device at the destination of a live migration we send the commands through control virtqueue. For a device to read CVQ it must have received the DRIVER_OK status bit.

However this opens a window where the device could start receiving packets in rx queue 0 before it receives the RSS configuration. To avoid that, we will not send vring_enable until all configuration is used by the device.

Add vhost-vdpa net vhost_vdpa_should_enable. Do not change the behavior in this commit, only introduce the op.

---
 net/vhost-vdpa.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
net/vhost-vdpa.c +10 -0
index e19ab063fa..536bab8613 100644
@@ -826,6 +826,15 @@ static const VhostShadowVirtqueueOps vhost_vdpa_net_svq_ops = {
     .avail_handler = vhost_vdpa_net_handle_ctrl_avail,
 };
 
+static bool vhost_vdpa_should_enable(const struct vhost_vdpa *v)
+{
+    return true;
+}
+
+static const VhostVDPAVirtIOOps vhost_vdpa_virtio_net_ops = {
+    .should_enable = vhost_vdpa_should_enable,
+};
+
 /**
  * Probe if CVQ is isolated
  *
@@ -949,6 +958,7 @@ static NetClientState *net_vhost_vdpa_init(NetClientState *peer,
     s->vhost_vdpa.shadow_vqs_enabled = svq;
     s->vhost_vdpa.iova_range = iova_range;
     s->vhost_vdpa.shadow_data = svq;
+    s->vhost_vdpa.virtio_ops = &vhost_vdpa_virtio_net_ops;
     if (queue_pair_index == 0) {
         vhost_vdpa_net_valid_svq_features(features,
                                           &s->vhost_vdpa.migration_blocker);
Signature · 1 line
2.39.3
#8 lore raw

RFC PATCH 5/6 vdpa: delay enable of data vqs

In reply to Eugenio Pérez

To restore the device at the destination of a live migration we send the commands through control virtqueue. For a device to read CVQ it must have received the DRIVER_OK status bit.

However this opens a window where the device could start receiving packets in rx queue 0 before it receives the RSS configuration. To avoid that, we do not send vring_enable until all configuration is used by the device.

---
 net/vhost-vdpa.c | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)
net/vhost-vdpa.c +24 -1
index 536bab8613..d5970e9f06 100644
@@ -739,6 +739,13 @@ static int vhost_vdpa_net_load(NetClientState *nc)
         return r;
     }
 
+    for (int i = 0; i < v->dev->vq_index; ++i) {
+        r = vhost_vdpa_set_vring_ready(v, i);
+        if (unlikely(r)) {
+            return r;
+        }
+    }
+
     return 0;
 }
 
@@ -826,9 +833,25 @@ static const VhostShadowVirtqueueOps vhost_vdpa_net_svq_ops = {
     .avail_handler = vhost_vdpa_net_handle_ctrl_avail,
 };
 
+/**
+ * Check if a vhost_vdpa device should enable before DRIVER_OK
+ *
+ * CVQ must always start first if we want to restore the state safely. Do not
+ * start data vqs if the device has CVQ.
+ */
 static bool vhost_vdpa_should_enable(const struct vhost_vdpa *v)
 {
-    return true;
+    struct vhost_dev *dev = v->dev;
+
+    if (!dev->vq_index_end % 2) {
+        /* vDPA device does not have CVQ */
+        return true;
+    }
+
+    /*
+     * We're evaluating CVQ, enable to send control cmds to load device state.
+     */
+    return dev->vq_index + 1 == dev->vq_index_end;
 }
 
 static const VhostVDPAVirtIOOps vhost_vdpa_virtio_net_ops = {
Signature · 1 line
2.39.3
#9 lore raw

RFC PATCH 6/6 vdpa: remove net cvq migration blocker

In reply to Eugenio Pérez

Now that we have add migration blockers if the device does not support all the needed features, remove the general blocker applied to all net devices with CVQ.

---
 net/vhost-vdpa.c | 12 ------------
 1 file changed, 12 deletions(-)
net/vhost-vdpa.c +0 -12
index d5970e9f06..5432b50498 100644
@@ -996,18 +996,6 @@ static NetClientState *net_vhost_vdpa_init(NetClientState *peer,
         s->vhost_vdpa.shadow_vq_ops = &vhost_vdpa_net_svq_ops;
         s->vhost_vdpa.shadow_vq_ops_opaque = s;
         s->cvq_isolated = cvq_isolated;
-
-        /*
-         * TODO: We cannot migrate devices with CVQ and no x-svq enabled as
-         * there is no way to set the device state (MAC, MQ, etc) before
-         * starting the datapath.
-         *
-         * Migration blocker ownership now belongs to s->vhost_vdpa.
-         */
-        if (!svq) {
-            error_setg(&s->vhost_vdpa.migration_blocker,
-                       "net vdpa cannot migrate with CVQ feature");
-        }
     }
     ret = vhost_vdpa_add(nc, (void *)&s->vhost_vdpa, queue_pair_index, nvqs);
     if (ret) {
Signature · 1 line
2.39.3
#10 lore raw

Re: [RFC PATCH 6/6] vdpa: remove net cvq migration blocker

Quoted 6 lines · Eugenio Pérez

Now that we have add migration blockers if the device does not support all the needed features, remove the general blocker applied to all net devices with CVQ.

I wonder what's the difference compared if we just start cvq first in vhost_net_start()?

Thanks

Quoted 30 lines
---
 net/vhost-vdpa.c | 12 ------------
 1 file changed, 12 deletions(-)
net/vhost-vdpa.c +0 -13
index d5970e9f06..5432b50498 100644
@@ -996,18 +996,6 @@ static NetClientState *net_vhost_vdpa_init(NetClientState *peer,
         s->vhost_vdpa.shadow_vq_ops = &vhost_vdpa_net_svq_ops;
         s->vhost_vdpa.shadow_vq_ops_opaque = s;
         s->cvq_isolated = cvq_isolated;
-
-        /*
-         * TODO: We cannot migrate devices with CVQ and no x-svq enabled as
-         * there is no way to set the device state (MAC, MQ, etc) before
-         * starting the datapath.
-         *
-         * Migration blocker ownership now belongs to s->vhost_vdpa.
-         */
-        if (!svq) {
-            error_setg(&s->vhost_vdpa.migration_blocker,
-                       "net vdpa cannot migrate with CVQ feature");
-        }
     }
     ret = vhost_vdpa_add(nc, (void *)&s->vhost_vdpa, queue_pair_index, nvqs);
     if (ret) {
--

2.39.3

#11 lore raw

Re: [RFC PATCH 6/6] vdpa: remove net cvq migration blocker

Quoted 12 lines · Jason Wang
Quoted 6 lines · Eugenio Pérez

Now that we have add migration blockers if the device does not support all the needed features, remove the general blocker applied to all net devices with CVQ.

I wonder what's the difference compared if we just start cvq first in vhost_net_start()?

That's interesting. It would complicate the for loop in vhost_net_start, but I think it's less complicated than should_start_op.

It would be something like moving from:

for (i = 0; i < nvhosts; i++) {
    if (i < data_queue_pairs) {
        peer = qemu_get_peer(ncs, i);
    } else {
        peer = qemu_get_peer(ncs, n->max_queue_pairs);
    }

    ...

    r = vhost_net_start_one(get_vhost_net(peer), dev);
    if (r < 0) {
        goto err_start;
    }
}

To:

for (i = 0; i < nvhosts; i++) {
    if (i == 0 && cvq) {
        peer = qemu_get_peer(ncs, n->max_queue_pairs);
    } else {
        peer = qemu_get_peer(ncs, i - cvq);
    }

    ...

    r = vhost_net_start_one(get_vhost_net(peer), dev);
    if (r < 0) {
        goto err_start;
    }
}

Is this what you have in mind?

Thanks!

Quoted 33 lines

Thanks

Quoted 30 lines
---
 net/vhost-vdpa.c | 12 ------------
 1 file changed, 12 deletions(-)
net/vhost-vdpa.c +0 -13
index d5970e9f06..5432b50498 100644
@@ -996,18 +996,6 @@ static NetClientState *net_vhost_vdpa_init(NetClientState *peer,
         s->vhost_vdpa.shadow_vq_ops = &vhost_vdpa_net_svq_ops;
         s->vhost_vdpa.shadow_vq_ops_opaque = s;
         s->cvq_isolated = cvq_isolated;
-
-        /*
-         * TODO: We cannot migrate devices with CVQ and no x-svq enabled as
-         * there is no way to set the device state (MAC, MQ, etc) before
-         * starting the datapath.
-         *
-         * Migration blocker ownership now belongs to s->vhost_vdpa.
-         */
-        if (!svq) {
-            error_setg(&s->vhost_vdpa.migration_blocker,
-                       "net vdpa cannot migrate with CVQ feature");
-        }
     }
     ret = vhost_vdpa_add(nc, (void *)&s->vhost_vdpa, queue_pair_index, nvqs);
     if (ret) {
--

2.39.3

#12 lore raw

Re: [RFC PATCH 6/6] vdpa: remove net cvq migration blocker

Quoted 53 lines · Eugenio Perez Martin
Quoted 12 lines · Jason Wang
Quoted 6 lines · Eugenio Pérez

Now that we have add migration blockers if the device does not support all the needed features, remove the general blocker applied to all net devices with CVQ.

I wonder what's the difference compared if we just start cvq first in vhost_net_start()?

That's interesting. It would complicate the for loop in vhost_net_start, but I think it's less complicated than should_start_op.

It would be something like moving from:

for (i = 0; i < nvhosts; i++) {
    if (i < data_queue_pairs) {
        peer = qemu_get_peer(ncs, i);
    } else {
        peer = qemu_get_peer(ncs, n->max_queue_pairs);
    }

    ...

    r = vhost_net_start_one(get_vhost_net(peer), dev);
    if (r < 0) {
        goto err_start;
    }
}

To:

for (i = 0; i < nvhosts; i++) {
    if (i == 0 && cvq) {
        peer = qemu_get_peer(ncs, n->max_queue_pairs);
    } else {
        peer = qemu_get_peer(ncs, i - cvq);
    }

    ...

    r = vhost_net_start_one(get_vhost_net(peer), dev);
    if (r < 0) {
        goto err_start;
    }
}
Quoted 2 lines

Is this what you have in mind?

Friendly ping.

Thanks!

#13 lore raw

Re: [RFC PATCH 6/6] vdpa: remove net cvq migration blocker

Quoted 50 lines · Eugenio Perez Martin
Quoted 47 lines · Eugenio Perez Martin
Quoted 12 lines · Jason Wang
Quoted 6 lines · Eugenio Pérez

Now that we have add migration blockers if the device does not support all the needed features, remove the general blocker applied to all net devices with CVQ.

I wonder what's the difference compared if we just start cvq first in vhost_net_start()?

That's interesting. It would complicate the for loop in vhost_net_start, but I think it's less complicated than should_start_op.

It would be something like moving from:

for (i = 0; i < nvhosts; i++) {
    if (i < data_queue_pairs) {
        peer = qemu_get_peer(ncs, i);
    } else {
        peer = qemu_get_peer(ncs, n->max_queue_pairs);
    }

    ...

    r = vhost_net_start_one(get_vhost_net(peer), dev);
    if (r < 0) {
        goto err_start;
    }
}

To:

for (i = 0; i < nvhosts; i++) {
    if (i == 0 && cvq) {
        peer = qemu_get_peer(ncs, n->max_queue_pairs);
    } else {
        peer = qemu_get_peer(ncs, i - cvq);
    }

    ...
Quoted 9 lines
Quoted 8 lines
    r = vhost_net_start_one(get_vhost_net(peer), dev);
    if (r < 0) {
        goto err_start;
    }
}

Is this what you have in mind?

Yes, something like this.

Thanks

Quoted 4 lines

Friendly ping.

Thanks!

#14 lore raw
In reply to Eugenio Pérez
Quoted 11 lines · Eugenio Pérez

At this moment the migration of net features that depends on CVQ is not possible, as there is no reliable way to restore the device state like mac address, number of enabled queues, etc to the destination. This is mainly caused because the device must only read CVQ, and process all the commands before resuming the dataplane.

This RFC lift that requirement, sending the VHOST_VDPA_SET_VRING_ENABLE ioctl for dataplane vqs only after the device has processed all commands. If this method is valid or not, or if it must be signalled by the parent driver somehow, is still under discussion. In case it is valid, this code allows testing the vDPA device for it.

And you plan to add the reset trick too in a future version?

Quoted 17 lines

Eugenio Pérez (6):
vdpa: export vhost_vdpa_set_vring_ready
vdpa: add should_enable op
vdpa: use virtio_ops->should_enable at vhost_vdpa_set_vrings_ready
vdpa: add stub vhost_vdpa_should_enable
vdpa: delay enable of data vqs
vdpa: remove net cvq migration blocker

 include/hw/virtio/vhost-vdpa.h |  9 +++++++
 hw/virtio/vhost-vdpa.c         | 33 +++++++++++++++++++------
 net/vhost-vdpa.c               | 45 +++++++++++++++++++++++++---------
 hw/virtio/trace-events         |  2 +-
 4 files changed, 68 insertions(+), 21 deletions(-)
Signature · 1 line
2.39.3
#15 lore raw
Quoted 8 lines · Michael S. Tsirkin
Quoted 6 lines · Eugenio Pérez

At this moment the migration of net features that depends on CVQ is not possible, as there is no reliable way to restore the device state like mac address, number of enabled queues, etc to the destination. This is mainly caused because the device must only read CVQ, and process all the commands before resuming the dataplane.

Quoted 8 lines
Quoted 5 lines

This RFC lift that requirement, sending the VHOST_VDPA_SET_VRING_ENABLE ioctl for dataplane vqs only after the device has processed all commands. If this method is valid or not, or if it must be signalled by the parent driver somehow, is still under discussion. In case it is valid, this code allows testing the vDPA device for it.

And you plan to add the reset trick too in a future version?

I'll try to come with a version today.

At the current state of vDPA devices, it will just serve to enable vp_vdpa to be the destination of a migration. But vp_vdpa cannot migrate the guest again, so it is a dead end. I can try to add _F_RING_RESET to vdpa_sim if that makes sense.

Anyway, as HW implement _F_RING_RESET, there should be no need to make more changes to qemu, so we should be good.

Quoted 18 lines
Quoted 17 lines

Eugenio Pérez (6):
vdpa: export vhost_vdpa_set_vring_ready
vdpa: add should_enable op
vdpa: use virtio_ops->should_enable at vhost_vdpa_set_vrings_ready
vdpa: add stub vhost_vdpa_should_enable
vdpa: delay enable of data vqs
vdpa: remove net cvq migration blocker

 include/hw/virtio/vhost-vdpa.h |  9 +++++++
 hw/virtio/vhost-vdpa.c         | 33 +++++++++++++++++++------
 net/vhost-vdpa.c               | 45 +++++++++++++++++++++++++---------
 hw/virtio/trace-events         |  2 +-
 4 files changed, 68 insertions(+), 21 deletions(-)

--
2.39.3